On Thu, 30 Jun 2016 18:18:02 +0000
"Ma, Peter" <peter.ma(a)intel.com> wrote:
> I would like to hear your thoughts about introducing an override to allow smaller data file sizes to be programmed. Thanks.
where would the data be put then? Should it be top-aligned like in all
x86 firmwares, should it be starting at 0 like in many other
applications (not sure about FPGA bit stream storage but I guess it is
one of these), or does it need to be put to another offset (like when
updating only parts of an image?
As you see the trivial override switch you suggest does not work that
well in many situations. However, we are of course aware of that and
there is a fix for this in the pipeline (referred to as layout patches)
that will allow to define names for address regions and use files
matching those address ranges in size to be flashed without a complete
Regarding your use case: is the bit stream image to be written to flash
always of a fixed size or does it depend on the code? I think that's
something we did not mind yet.
Kind regards/Mit freundlichen Grüßen, Stefan Tauner